Blackwell Science, LtdOxford, UKVHEValue in Health1098-30152003 ISPOR6Supplement 1S10S19Original ArticleThe Drug Budget Silo Mentality: The French CaseLe Pen
Volume 6 • Supplement 1 • 2003
V A L U E I N H E A L T H
The Drug Budget Silo Mentality: The French Case
Claude Le Pen, PhD
University of Paris-Dauphine, Paris, France
ABST R ACT
Objectives: The objectives of this study were to give a
review of the complex system of budgetary constraints to
which the French health-care system has been committed
since 1996 and to evaluate the consequences on drug policy and on efficient use of pharmaceuticals.
Methods: Literature review, legal texts analysis, and
interviews with policy makers and companies managers
were performed.
Results: The budgeting process applies to health insurance expenditures as a whole, but also to each of its
components, especially hospital expenditures and pharmaceutical expenditures. Because the targets are set by
reference to the gross domestic product growth while
health-care expenditure is driven by demographic factors,
technology, and expectations, there is inevitably a gap
between the top-down budget and the bottom-up cost
pressures. The pharmaceutical budget is achieved by a
payback system that “taxes” companies when the growth
target for aggregated pharmaceutical expenditure is
exceeded. The government is now seeking to set more
realistic overall global budgets and for pharmaceuticals in
particular. It is also encouraging generics, delisting from
reimbursement drugs of limited therapeutic value, making
a special budget for new drug purchases available to hospitals, and replacing price control for innovative products
with a more selective process of intervention in the expectation that companies will seek to price at a European
level.
Conclusion: The budgetary system produces a perverse
incentive for companies to heavily promote new products
in the knowledge that the budget overruns will be spread
across all companies, as well as lacking incentives for using
pharmaceuticals efficiently. Although the new drug policy
will increase the efficiency of pharmaceutical expenditure,
it is not apparent that they change the poor incentives facing doctors, hospitals, and insurers to use pharmaceuticals
cost-effectively to achieve the optimal gain in health care.
They will not remove “silo budgeting” at the national level
for pharmaceuticals, which inhibits the efficient substitution of drug therapy for hospital treatment.
Keywords: France, drug policy, health-care budgets,
health policy, health insurance, health-care expenditure.
Introduction
to the OECD average. This is a common feature of
the southern European countries. Total drug consumption in France was about €27.3 billion in
2000, and the total turnover of the pharmaceutical
industry, €31.5 billion [2]. The average per-capita
expense is one of the largest among western countries (€448 in 2000).
The French drug market is a low-price and highquantity market. A recent study computed a series
of price and quantity indexes, from IMS Health
1997 data, comparing the United States to different
European countries for a set of identical products:
the resulting (Laspeyre) quantity index was more
than 2 for France compared to the United States,
whereas the price index was 0.5. The corresponding
statistics are 1 and 0.9 for the United Kingdom and
1.1 and 0.9 for Germany [3]. The market share of
homeopathic and herbal products is high compared
to other countries. The market for plant extracts
France is known as being one of the western countries that devotes a large part of its gross domestic
product (GDP) to health care. According to OECD
health data, the ratio of total health expenditure to
GDP was 9.5% in 2000, which was beaten only by
the United States (13%) and Germany (10.6%).
Various economic as well as institutional factors
explain this position. France in fact combines a high
per-capita GDP and a social health insurance system, which are both factors generally associated
with high health expenditure rates (Fig. 1) [1].
The share of pharmaceuticals in total health care
is was 20% in 2000, which was also high compared
Address correspondence to: Claude Le Pen, LEGOS,
Université Paris-Dauphine, Paris 75015, France. E-mail:
[email protected]
© ISPOR
1098-3015/03/$15.00/S10
S10–S19
S10
S11
The Drug Budget Silo Mentality: The French Case
Complementary Insurance Funds
"Mutuelles"
Nonprofit
Insurance Cies
35%
35%
Commercial
Insurance Cies
Dental care
Optical care
30%
Pharmaceuticals
20%
Doctors' fees
5%
Hospitals
4%
60%
20%
Public Health Insurance Funds
75%
CNAM (salaried workers)
MSA (farmers)
CANAM (independent business)
20%
92%
Salary/riskbased
premiums
(100%)
Salary-based
"contribution"
(60%)
Special income tax
(40%)
4%
Population
Figure 1 The French health-care system organization. The figure
depicts the main financial flows in the French health-care system.
Pharmaceuticals, for instance, are financed 60% from public health
insurance funds, 20% from complementary insurance, and 20% out of
pocket. Public funds are financed 60% from social contribution and
40% from a special income tax.
products, mainly produced by small family
companies, is also high. Finally, the generic drugs
market share is about 3.5% (2001) of the total market, which is low compared to the United Kingdom,
Germany, or the United States. In fact, prescription
and reimbursement conditions create little incentives for generic consumption in France.
In France, as in most other western countries,
drug costs have continuously risen more rapidly
(nominal growth rate was 8% in 2001) than the
costs of other health-care sectors (total health-care
expenditure growth rate was about 5.7% in 2001).
This growth is ascribable neither to the quantities of
consumed drugs, which grow slightly (+1.6% in
2000) nor to the prices of the existing drugs, which
rather tends to lower (−0.7% in 2000). It is
explained primarily by the launching of new innovating products that replace the old ones and that,
for various reasons, are often much more expensive.
Health-Care Financing
Health Funds
France has a social insurance system in which health
insurance is mainly provided through sickness funds
independent from the state and financed through
social contributions, equally shared between
employers and employees. Agricultural workers,
salaried workers, farmers, and independent professions, such as self-employed workers, shopkeepers,
and lawyers, have their own fund (respectively the
MSA, “Mutualité Sociale Agricole”; and the
CANAM, “Caisse d’Assurance-Maladie des Professions Indépendantes”), separate from the general
wage-earners sickness fund (The CNAMTS,
“Caisse Nationale d’Assurance-Maladie des Travailleurs Salariés”). This latter reassembles nevertheless approximately 80% of the total population.
Apart from these three main funds, other minor
funds cover some specific populations, for instance,
the miners.
One of the major changes that have recently
occurred in the French public health insurance system is the shift of a substantial part of the financing
from social insurance contribution to taxes. A special tax, named “generalized social contribution”
(CSG), was introduced in 1990 to tackle rising
social protection expenses. The tax base constitutes
all wage and nonwage incomes, including real estate
income and financial investment income. In 1998,
this tax replaced the employees’ social insurance
contribution, which was almost completely abolished. The initial 2.4% CSG flat rate was consequently raised to 7.5% of gross income. As a
consequence, whereas in 1997 fiscal taxes represented less than 10% of the financing of public
health insurance, they now account for approximately 40% (Table 1). This change had political
consequences because it reinforced the role of the
state in the management of the health-care system at
the expense of employers’ and employees’ representatives.
Complementary Health Insurance
In France, most of the population holds an optional
private health insurance coverage, which “complements” the mandatory public coverage; that is, it
reimburses totally or partly the public insurance
copayment. This two-level system of health insurance (public mandatory plus private optional) is
unique to France. Three types of institutions
provide complementary health insurance, the
“Mutuelles”, the “Institutions de Prévoyance,” and
most of the commercial insurance companies.
“Mutuelles” are the heirs of the 19th century
mutual assistance societies spontaneously organized
by workers on a professional basis. These nonprofit
institutions survived the creation of the public
health insurance (“Sécurité Sociale”) in 1945 and
specialized in delivering complementary health
insurance to their members. They are especially
strong in the public sector. The “institutions de
prévoyance” are private nonprofit institutions managed by both employers and employees. Initially
created to provide supplementary retirement pensions in the private sector, they have extended their
activity to complementary health insurance. They
are dominant in the private sector through collec-
Le Pen
S12
Table 1 The financing of public health insurance in France (2000)
Expenditures
Amount (millions €)
Inpatient care
Public hospitals
Private hospitals
Outpatient care
Physician services
Dental services
Pharmaceutical goods
Total expenditures on medical
services and goods
55,264
42,729
10,679
31,861
15,324
6,430
25,069
120,640
%
Social security
45.8
35.4
8.9
26.4
12.7
5.3
20.8
100.0
91.1
92.3
84.8
67.0
75.6
35.3
63.6
76.6
tive contracts, partly paid by employers, which are
considered as an employment fringe benefit. In both
cases, insurance premiums are based on gross
income and not on the insured people’s health risk.
Commercial insurance companies have two markets. They compete with “institutions de prévoyance” to provide collective contracts to private
sector wage earners and their employers and they
sell individual contract to self-employed or unemployed people. In the latter case, premiums are riskdependent. In the past, the existence of a private
level of complementary health insurance has limited
the impact of government cost-sharing strategies to
contain health-care cost rise.
Universal Coverage
Public health insurance covers practically all residents in France (99.5%). This is mainly through
their professional status, which includes the scheme
for the unemployed. The 1.8% of the population
who are without a definite professional affiliation,
for instance, divorced nonworking women, have
access to public insurance through a direct personal
application called “residential status.” The population holding a complementary health insurance is
now estimated at 92%. This number substantially
grew under law by the “Universal Sickness Coverage Law,” passed in 1999, which entitled poor people to free complementary coverage. In July 2002,
3.5 million people benefited from this measure.
The basic benefit package is extensive and
includes a full range of goods and services. Hydrotherapy, homeopathy, transportation, and of
course, pharmaceuticals are all included. Inclusion
or exclusion of a service is a state decision, which
does not depend on the views of the sickness funds.
The package covered by the complementary institutions is generally the same as that of public insurance with some, albeit rare, exceptions. Some
complementary institutions, especially commercial
companies, have started to cover services that are
Financing by source of funds (%)
Complementary insurance
Households
3.7
2.3
10.1
21.4
20.4
35.9
18.6
12.4
5.2
5.4
5.3
11.7
4.0
28.7
17.9
11.1
not reimbursed by public insurance, for instance the
antiflu drug Relenza® (GlaxoSmithKline, Research
Triangle Park, NC). This tendency will certainly
grow in importance in years to come.
On average, the three public sickness funds cover
76.6% of total health-care expenditure. A total of
12.4% is financed through complementary health
insurance institutions and the remaining 11% is
out-of-pocket spending (Table 2). The division of
financing as between public insurance, complementary health insurance institutions, and out-ofpocket spending by patients varies according to the
type of care. The share of public insurance is above
90% for hospital care, whereas it is only 35% for
dental care. For pharmaceuticals, the relative shares
of public insurance, complementary insurance, and
out-of-pocket payment are, respectively, 64, 19, and
18%.
Health-Care Budgeting: Public Hospitals
Since the middle of the 1980s, hospital care in
France has been strictly regulated by budgets, annually set by the government. Initially, the rate of
annual budget increase was the same for all public
hospitals and computed, roughly, from a combina-
Table 2 The financing of medical consumption in France
(2000)
1997
Taxes
General social tax (CSG)
Other taxes
Pharmaceutical industry
Social contribution
Employers (private sector)
Employees (private sector)
Other employees
Other
Total
7.5
5.3
2.2
—
73.1
42.8
19.7
10.6
0.7
81.3
Billion euros
%
2001
9.2
6.5
2.7
—
89.9
52.6
24.2
13.0
—
100.0
39.8
36.9
2.9
0.5
59.5
47.7
2.9
8.9
0.2
99.5
%
40.0
37.1
2.9
0.5
59.8
47.9
2.9
8.9
0.2
100.0
S13
The Drug Budget Silo Mentality: The French Case
tion of expected economic growth and the public
sector wage index; staff expenses account for about
70% of hospital budgets [4]. Later, in the 1990s,
productivity indexes were computed for each public
hospital, as a by-product of the introduction of a
diagnosis related groups (DRG) system, the French
version being called PMSI, which stands for “Programme de Médicalization des Systèmes d’Information.” Costs per DRG are available for each public
hospital but they are not at present used to finance
hospitals. They are used, instead, to make hospital
budget comparisons on a case-mix-adjusted basis.
More “productive” hospitals are thus likely to negotiate higher budgets. Less productive hospitals are
asked to make productivity gains before they can
expect to benefit from budget increases.
Since 1996, budgets have been determined by
regional state agencies called ARHs (“Agences
Régionale d’Hospitalization”), but they are still
paid by health insurance. So the state determines the
budgets but does not pay the hospitals, leaving the
health insurers paying for but not determining
the budgets. This situation creates tensions between
the state and the health insurers, especially in 1999
when health insurance claimed a greater control on
hospitals expenditures. This was refused by the government, which wanted to keep a direct control on
public hospitals. The ARHs also have a mission to
restructure the hospital beds available at the regional level. Hospitals are requested to produce 5year development plans and ARHs may encourage
financially some development projects that seem
consistent with their regional hospitalization policy.
Globally, nevertheless, this budget system resulted
in a significant slowdown in annual cost increase
during the 1980s and the 1990s. Despite the introduction of a productivity measure, budgets acted
more and more as severe and effective rationing
devices. Hospitals have some difficulties buying
some innovative expensive drugs such as the new
TNF-α inhibitors in rheumatoid arthritis, which are
in France restricted to hospital use only. Although
there are officially no waiting lists, some surgical
interventions must be delayed at the end of the year.
Shortages for some categories of medical and nonmedical personnel appear. Most hospitals are simply not in a position to comply with the recent
reducing work time law, which limits to 35 hours
the duration of the weekly working time. The
resulting crisis burst out in summer 2002 and the
future of public hospitals has become a sensitive
political issue. This induced the new ministry of
health to prepare a €1 billion emergency plan for
public hospitalization (2003–2007).
Total Health-Care Budgeting: Process
and Issues
Budgeting Total Reimbursed Health-Care Expenditures
In 1996 the government made a comprehensive
attempt to dramatically reduce the Social Security
deficit, which was at that time about €9 billion, as a
result of slow economic growth in 1994 and 1995.
Public deficits had to be quickly reduced to qualify
the country to join the Euro. The “Juppé Plan,”
named after the French Prime Minister, established
a sophisticated system of budgets covering all healthcare sectors.
At the top of this system the Parliament voted a
National Objective for Health Insurance Expenditures (“Objectif National d’Assurance-Maladie,”
ONDAM). This objective was thus set at €112.8
billion in December 2001 for the year 2002 and at
€136.3 billion in December 2002 for year 2003. No
formal justification has ever been produced for this
kind of figure. It mainly depends on macroeconomic
variables such as the predicted rate of GDP growth
or the public deficit. In a second step, this budget
is broken down by the government (not the
Parliament) into four subbudgets, respectively, for
short-term public hospital (€43.2 billion in 2001),
long-term hospitals (called “medicosocial institutions”, €7.9 billion), private hospitals (€7 billion),
and ambulatory care (€51.5 billion). Short-term,
long-term, and private hospital budgets are directly
managed by the state through the regional agencies
as described previously. Within the ambulatory
budget, the part relating to pharmaceuticals is also
directly managed by the state according to the procedure that will be described in the next section.
Public health insurance, mainly the CNAMTS, is
entrusted with managing the rest of the ambulatory
care budget, which essentially relates to doctors’
and other health-care professionals’ (HCP) fees.
To reconcile between the tight financial budgetary envelope and the strong growth in consumption volume, the government mainly exerts a
downward pressure on prices in the ambulatory
sector and rations financial resources to the hospital sector. Because this is not enough, there is a
third adjustment variable: the budget overrun.
Since the implementation of this mechanism in
1996, the government has never been able to contain health insurance expenditures within the limit
it fixed itself (Table 3). At the time it is voted, it is
already known that the objective will not be
achieved. The solemn vote of a fictive objective is
more and more perceived as a denial of the parliamentary mission.
Le Pen
S14
Table 3 Rate of increase of reimbursed health-care
expenditures
Years
1997–1998
1998–1999
1999–2000
2000–2001
2001–2002
Budgeted (%)
Realized (%)
2.3
2.6
2.5
3.5
3.9
4.0
2.8
4.7
5.8
7.2
Budget or Target?
The status of the total budget for health care is
highly ambiguous: is it a financial budget which
must be respected by the various stakeholders,
mainly hospital managers, insurance funds managers, and health-care professionals, or a simple target,
acting as a political signal from the government, setting the level of reimbursed health-care expenditure
the nation can afford at a constant tax and social
contributions rate? This has never been clarified. In
contrast, as already mentioned, the total budget is
set mainly by reference to macroeconomic variables.
No demand variables such as demographic growth,
population aging, technical progress, and changes in
sociological attitudes toward health and wellness
are taken into account or even considered. So the
stakeholders do not feel committed in fulfilling the
target. But on the other hand, each year, health
insurance and HCPs’ representatives are urged by
the government to find ways to limit the gap between the objective and the reality. Every 3 months,
a round of negotiations is started between HCP
unions and the public sickness funds. The most
common decision is a negative measure: not to raise
official tariffs. Thus general practitioners’ (GPs’) fee
for an ordinary consultation has long been frozen at
its 1996 level, namely, €16.8 (110 French francs). In
September 2001, the claim for a raise to €20 was
rejected both by the government and by the sickness
funds, and GPs came out on strike for all emergency
consultations and visits from November 2001 to
June 2002; all night and weekend calls were redirected toward overloaded hospitals. Public opinion
supported the doctors’ action, and one of the first
decisions of the new government, elected in June
2002, was to give satisfaction to the GPs, imputing
responsibility for the crisis to the previous government. This kind of situation and crisis is frequent
and happened for the same reasons with each of the
HCPs, including dentists, nurses, and midwifes.
HCPs believe that the budgeting system is a mechanism aimed at making them guilty. Each time the
budget is overrun, each year in practice, HCPs are
held responsible for the expenditures escalation and
the deficit, and there is a call for corrective action.
The climate has deteriorated between HPCs and the
government, and public opinion becomes more and
more anxious about the future of what has long been
considered a national pride, ranked by the WHO as
the best performing health-care systems in the world
[5]. Health-care management, which has so long
been considered as a purely technical cost-containment problem, has become a politically sensitive
issue, at stake being the survival of a comfortable
and balanced public–private system.
Pharmaceutical Budgeting
As seen previously, the Parliament only votes the
global budget for reimbursed health-care expenditures. Subbudgets for each type of care, short-term
and long-term hospitals, HCPs’ fees, pharmaceuticals, etc., are set directly by the government. In particular, drugs are submitted to a rather complicated
budgeting system, which works as follows:
1. First the government issues unilaterally a rate of
“normal” growth for reimbursed pharmaceuticals, called the “k rate.” Again, no justification
is given for this rate, which was set at 3% in
2001, a lower rate than the general rate for all
health-care expenditures (3.8%).
2. If the actual growth is above the planned
growth, the whole industry must pay back to
health insurance funds a part of the difference,
ranging from 50% to 70% according the magnitude of the difference, “the safeguard clause.”
3. The global discount must then be divided
among companies. Here, there is a choice for
each individual company: whether or not to
sign a “convention” with the state (the Drug
Price Committee). The method of computation
of the individual discounts depends on this
choice.
4. If a company chooses not to sign a convention,
an individual discount is computed according to a mechanistic formula (the “safeguard
procedure”), proportionally to its turnover, to
the growth of the turnover, and to the amount
of promotional expense. Only sales of reimbursed prescription products in France on
the pharmacy market, excluding over the
counter (OTC), exports, and hospital sales, are
considered.
5. If, in contrast, a company chooses to sign an
agreement, the “conventional procedure” applies. It is far more complicated. First, the Drug
Price Committee attributes to each of the 122
S15
The Drug Budget Silo Mentality: The French Case
therapeutic classes under consideration a “permitted” rate of increase. Again no justification
will be provided and the rates are just discretionary. Then, the Drug Price Committee computes the expenses overrun for each of the class
and for the aggregated market. The companies
are charged collectively 35% of the expenses
excess (2001 rate). The amount is broken down
according to companies’ turnover and growth
of turnover; promotional expenditures are no
longer taken into account. Generally speaking,
the amount of money, which is recovered this
way, is not enough. So a last negotiation takes
place, company by company, to obtain a supplementary discount. This last negotiation is
not formulaic and plays the role of “gray area”
allowing some form of discrimination among
companies.
Companies have developed a certain skill to
“play” with these procedures. Simulations have
shown that most of the time the best solution was to
sign the agreement and to engage the complicated
conventional procedure. In fact, the discount under
the conventional procedure is approximately 75%
to 80% of what it would have been under the formulaic application of the safeguard procedure and
100% of companies signed a convention in 2001.
In 2001, the total discount paid by the industry
was €526 million. The industry also paid substantial special taxes: €183 million in promotional
taxes, €106 million in direct sales taxes, and €89
million in contributions to the Drug Agency (this
tax applies to direct sales to retail pharmacists,
which do not pass through wholesalers. It creates an
incentive to use the conventional “long” distribution track as a compensation for the legal obligations of wholesalers to have in stock at least 90% of
existing drugs, to maintain a stock equivalent to 2
weeks of drug consumption and to deliver any retail
pharmacy within 24 hours). In 2001, there was also
a price cut, targeted on the best-selling drugs. The
total amount paid by the industry, directly or indirectly through price cut (amounting to approximately €1.10 billion), apart from normal business
taxes. This is approximately the value of the net
after tax profit of the whole industry, including
export, hospital sales, OTC, etc.
Pharmaceutical Budgeting and
Expenses Regulation
We now must examine how this complex budgeting
procedure affects both the regulation of pharmaceu-
ticals expenditures and the use of drugs by prescribers and consumers.
Reimbursement
In France, a system of positive list prevails for
drug reimbursement. The Transparency Commission within the Drug Agency proposes a listing of a
new compound. A drug can be listed if its “Medical
Service Rendered” (MSR) is deemed “sufficient”
with regard to a list of criteria formally set by a
recent decree in October 1999: 1) drug efficacy and
safety; 2) severity of the disease; 3) place in the therapeutic strategy; 4) existence of alternative treatments; and 5) public health value.
In fact, as proved by a statistical analysis performed on the results of the MSR evaluation of
1453 drugs in five therapeutic areas, only two criteria—efficacy and seriousness of the disease—suffice to very largely explain the MSR classification.
The other criteria contribute little added value.
The point here is that there is no relationship
between the budgeting system and the reimbursement procedure. The pharmaceutical budget is fixed
independently of the number and the nature of the
new products that are agreed for reimbursement.
Under these conditions, it is practically unavoidable
that the budget is systematically overcome, especially when breakthrough innovative products are
launched. Besides, in that case, there is a strange
and paradoxical effect. Whereas the benefit of marketing of new innovative product accrues to a single
company, all the industry is jointly liable to pay the
discount, which results from the new drug. It is thus
in the interest of this latter company to promote and
advertise heavily its breakthrough product, as the
payback payment is partly socialized.
Pricing
Drug price control has long been the main regulation tool for pharmaceutical expenditures. In fact,
only reimbursable drugs on the pharmacy market
are price regulated. There is no price regulation for
OTC drugs and for hospital-only drugs (in the latter
case, prices are discussed freely between the industry and the hospital pharmacist, who is often the
chairman of the local Prescription and Therapeutic
Committee). A state committee sets regulated prices
(the Drug Price Committee within the Ministry of
Health) after a price negotiation with applicant
companies. There are no formal rules and the open
negotiation is based on various criteria such as the
innovative nature of the product, the benefit to
patients, the projected market size, the competitor
domestic prices, and the foreign prices. Unlike some
Le Pen
S16
other countries, there is not in France a formal comparison with prices from a set of defined countries.
Nevertheless, there is a clear tendency on the part of
companies to align innovative product prices with
the “European price” to prevent parallel trade. The
Drug Price Committee, especially for innovative
products, generally accepts this argument. In fact,
because of the budgeting and payback system, price
regulation appears to be less necessary as a costcontainment tool. A clear tendency of the price
committee, at least for innovative products, is to
rely less on cost-containment through price regulation and more on raising additional resources
through the discount system. In that sense, there is
a clear impact of the budgeting policy on the pricing
policy.
Drug Utilization
The budgeting system operates only at the central
level and only drug companies are concerned by the
discount payment. There are thus no individual prescribers’ budgets and no financial incentives for
physicians to reduce or to improve their prescribing
practices.
In 1993 to 1994 yet, initiatives were taken at the
national level, with the help and support of the
main physicians’ union, to develop a set of mandatory clinical practice guidelines. These guidelines
were formulated in negative terms (for instance,
“there is not cause to use together two anticholesterol drugs, unless. . .”). Later on, some more traditional positive clinical recommendations were
issued either by the Drug Agency or by the Agency
for Health Care Technology Assessment (ANAES).
Although these recommendations had a great
impact on physicians, this strategy aiming at
improving the quality of drug utilization became
more or less obsolete.
A similarly timid policy was conducted with
respect to generic drugs. In 1999, pharmacists were
given a limited substitution rights for true generic
products, which are officially listed by the Drug
Agency. The commercial profit margin on generics
was maintained at the same level as that of the
reference product—although the generic price is
lower—to neutralize the pharmacists’ negative
incentives. But, patients, doctors, and pharmacists
still face no positive incentives to consume, prescribe, or deliver generic drugs. Recently, in June
2002, the physicians’ unions agreed to develop
generic prescribing as a counterpart of the agreement on fees revision, but no quantitative objectives
and no sanctions were set out in the agreement.
There is therefore nothing surprising about the very
low market share of generic drugs in France (less
than 4%). In fact, the budget policy has progressively occupied all the government’s attention, since
it was launched in 1996, at the expense of more
qualitative policy measures aiming at improving the
quality and efficiency of drug use.
Pharmacoeconomics and Pharmacoepidemiologic
Studies
Health economic evaluation studies applied to
pharmaceuticals (“pharmacoeconomic” studies) developed in France as in some other countries in the
mid-1980s. But it remained an academic discipline
or a consultancy product for the industry. Pharmacoeconomic studies have a little role to play in a
macroeconomic budgeting process. Public authorities in France were always reluctant to use it in
public decision making about reimbursement or
pricing. The reasons certainly relate, at least partly,
to the alleged “lack of credibility” of pharmacoeconomics. But the fact that it proves to be a poor costcontainment tool is probably a more decisive
explaining factor. Nevertheless, submitted studies
are examined by a group of independent experts
within the Drug Agency. Public authority representatives were also part of the elaboration of the
“Guidelines for Economic Evaluation in Health
Care,” undertaken under the auspices of the French
“Collège des Economistes de la Santé”.
A recent tendency from public authorities is to
encourage pharmacoepidemiologic studies, once
products are on the market. The reasons are twofold: first, to monitor physicians’ prescribing practice, and then to check new products’ effectiveness
in “real-life” conditions. This last objective,
although it is highly desirable, raises some very difficult methodological issues; especially the traditional question of knowing whether observed real
life effects are to be attributed to the drug itself or to
the use of the drug, which depends on other actors,
especially prescribers. The status of theses studies
remain somewhat obscure especially the relationship to the regulation process.
Breaking with the Budgeting Logic
In France, health-care policy was progressively
reduced to be nothing else but a financial budgeting
policy. This was especially obvious during the period
1996 to 2001. Each year the discussions of the
“Social Security Financing Bill,” which leads to the
vote of the ONDAM, mainly consists of setting
The Drug Budget Silo Mentality: The French Case
the targets and looking for repressive means to
achieve them.
All other health policy strategies have been more
or less given up. Cost sharing has been reduced
with the extension of complementary health insurance and exemption of copayments for a large part
of the population (8.5%). Payment for more than
two-thirds of pharmaceuticals acquired by patients
is made directly to pharmacists by the health insurance via electronic bank transfers, so that patients
have the feeling that pharmaceuticals are nothing
but a free good. The idea of implementing a dose
of competition between sickness funds was not
even discussed. More important, looking for financial savings through a quality improvement policy
was a strategy, which was pursued during a short
period (1993–1994) but quickly abandoned because it provides benefits only in the long term,
even if the effect on the system efficiency can be
immediate.
The emphasis put on the macroeconomic budgeting policy was in fact consistent with the raising of
the role of state regulation at the expenses of sickness funds, as witnessed by the “Juppé Plan” of
1996. Greater state control and the rise of budgeting strategies were in fact closely connected.
The main drawback with this budgetary system,
apart of its complexity, is the lack of incentives for
doctors and managers to use pharmaceuticals efficiently. Only the expense side of pharmaceutical
innovations is considered, irrespective to the benefits to patients and society. The silo effect of having
a separate budget for pharmaceuticals and hospitals
forbids a global computation of the financial effect
of drug innovation on total health-care expenditures when an ambulatory treatment results in
financial savings in the hospital sector. This was
especially the case when antiretroviral therapies for
HIV infection were made available to patients in
pharmacies in 1997. Before this date, these treatments were restricted to hospitals only. Now their
costs weigh on the pharmaceutical budget, whereas
substantial benefits in term of reduction of hospital
stays still accrue to the hospital sector [6,7]. No global view of the financial impact of the therapies is
thus available to decision maker.
Progressing Toward a New Drug Policy?
It has been progressively acknowledged that the
budgeting policy, as applied during the period 1996
to 2001, was not effective. The new government installed in June 2002, after Jacques Chirac was re-
S17
elected as president, has started to change healthcare policy. The new “Health Care Financing Bill
for 2003” gave the opportunity to draw new perspectives, especially in the pharmaceutical sector.
The following measures were announced:
1. A more realistic financial target for total health
care (5.4%) as well as for pharmaceuticals (k
rate set at 4% compared to 3% the previous
year).
2. The introduction of a reference pricing system
for all molecules with generic competition. For
the first time, market price and reimbursement
price will be disconnected, although competition will certainly tend to align prices, at least
in a majority of cases (curiously, this measure
was presented as a generic promotion policy,
whereas normally, it would rather advantage
the branded version of the molecules: prescribing or delivering a generic drug is strictly of no
interest as soon as the reference drug is sold and
reimbursed at the same rate).
3. The de-reimbursement of low-therapeuticvalue drugs within the next 3 years. This was
the most spectacular and the most courageous
measure, as French family companies market
most of these products. This measure concerns
600 compounds with a total turnover estimated
to €1 billion. Taking into account the distribution cost and the reimbursement rate (35%),
this measure should provide €450 million in
savings.
4. The adoption of a “price notification procedure” for innovative drugs. In this system, companies will be allowed to market innovative
drugs at their own price. Government will have
then a time period to oppose if price is exceedingly high. After that period without a reaction
from the state, the price becomes definite. This
measure is intended to discourage parallel trade
as it is expected that drug companies will
spontaneously price their new products at
the “European level”. Another objective is to
reduce marketing delays, which are excessively
long in France, because of a tedious reimbursement and price negotiation process. The scope
of that measure (all the new products or only
the innovative ones?), the time limit for state intervention, and the definition of an “excessive”
price are still to be negotiated.
5. Finally, a €350 million special budget for innovative drug acquisition has been allocated to
public hospitals.
S18
These measures have been well received by public
opinion and by the innovative part of the pharmaceutical industry. In fact, their adoption marks a
decline in the political influence of French family
drug companies that have long been dominant.
New Perspectives
In a longer term, it can be easily foreseen that the
political debate around the health-care system will
bear on a limited number of important topics.
First, the meaning of fixing a global budget for
health care is to be clarified. A state commission has
been appointed to make propositions for a redefinition of financial targets. Physicians unions push the
idea of a “scientific” definition of the total budget
based on the identification and quantification of
health “needs.” There is a form of illusion to believe
that the solution lies in that direction, given the
well-known ambiguity about the notion of “need.”
As Allan Williams pointed out some years ago, a
certain medical rhetoric uses in fact the concept of
need as a supply concept, leading to a form of circular reasoning: most of the time it is possible to
invoke need to justify any actual level of expenses,
leaving thus a limited place for rationalization and
improvement of the system management [8]. On the
other side, a bureaucratic process for budget determination inevitably leads to tensions, conflicts, and
rationing. In the French context, a more realistic
way may be to consider the budgeting process as a
collective negotiation, including all the stakeholders
(HCPs, the pharmaceutical industry, and the health
funds managers) about what is desirable and what
is achievable in term of health-care growth. In fact,
if stakeholders are granted responsibility for the
good execution of the budget, they should be associated to its determination.
Second, while it is legitimate that the government
decides what amount of public resources is to be
dedicated to health care, the breakdown into separate budget provides numerous perverse effects. Efficient use of health-care resources is prevented simply
because cost and benefits are not necessarily located
in the same budget. This is especially true for pharmaceuticals. Many drug innovations result in diminished hospital costs. Recently, for instance, we
showed that the number of trabeculum surgeries
(trabeculectomies and trabeculotomies) performed
in France was reduced by 16% (from 19,233 to
16,133 interventions per year) between 1997 and
1999. During the same period, the number of
patients treated with a new glaucoma drug increased
from 476,000 to 760,000 equivalent patients per
year. This increase is because of the launch of three
Le Pen
new glaucoma drugs. Development of fistulizing or
sclerotic surgery may explain part of the reduction
observed in private clinics, but not in public hospitals. In fact there is no other explanation for the dramatic reduction of trabecular surgery in hospitals
than new medical treatments, mostly latanoprost
and brimonidine providing better intraocular pressure control and delaying or avoiding surgery [9].
This kind of benefit of drug treatment cannot be captured in a separate budget structure.
Third, while under above conditions a total
budget may be useful to explicit health-care policy
choices, it cannot substitute all other policy measures, especially those that aim at improving the
quality and efficiency of health care, especially prescribed medicines. This strategy has been neglected
until now, except for a short period in the mid1990s, although it has a great potential. Government should also look for ways to make patients
more responsible about their health-care consumption, without necessarily increasing copayment. Patients’ education programs may provide a
solution.
Finally, if the budget resulting from a global
negotiation with stakeholders results in expenses
that exceed financial resources, new sources of
funding should be found or the benefit package
should be reduced. Public authorities and public
opinion must accept the idea that growth in healthcare expenditures at a rate faster than that of GDP
is a fact of life, at least for the present times. It has
long been presented as an institutional dysfunction
attributable to any one or more of an excess of
insurance coverage, ineffective state controls, or
manipulating corporatist strategies from HCPs. But
such a universal phenomenon throughout the developed world cannot be reduced to French circumstantial and institutional causes. It is a basic
characteristic of modern western democracies that
should be accepted as such rather than denied. Of
course, health-care expenditure cannot grow faster
than GDP forever. The history of household consumption shows that some goods may be superior
goods during a period of time and then become inferior goods in the continuous course of the economic
and social development. But the time horizon for
the health-care budgeting process is much shorter.
In the short to medium term, although the French
government has announced that it would not raise
taxes and social contributions, it is hard to imagine
that it can successfully face the current difficult situation without finding new sources of funding. A
reform of the employers’ contributions and an
extension of the new CSG tax can be envisaged,
S19
The Drug Budget Silo Mentality: The French Case
although this would further reinforce the leading
role of the state.
In the French context, it is also clearly possible to
think about reconsideration of the public health
insurance benefit package and more precisely to the
search for a new balance between public health
insurance and complementary health insurance
institutions. The debate is already launched around
the idea of having universal public coverage limited
to a rather large package of prioritized goods and
services, with the rest being financed through complementary health insurance. The fact that more
than 90% of the population is now covered by such
a complementary insurance limits the potential
equity risk that underlies such a construct.
In that perspective, a global budget can be a useful tool for rationalizing health-care expenditures
and improving transparency in collective choices. It
cannot and should not be limited to what it is now:
a raw bureaucratic management tool imposed unilaterally to HCPs and patients.
I thank Adrian Towse, Bengt Jonsson, and Mike Drummond for helpful comments on an earlier draft. I am, of
course, responsible for all remaining mistakes, errors, and
omissions.
References
1 Gerdtham UG, Sogaard J, Andersson F, Jonsson B.
An econometric analysis of health care expenditure:
a cross-section study of the OECD countries. J
Health Econ 1992;11:63–84.
2 Eco Santé 2002. Paris: CREDES, 2002.
3 Danzon PM, Chao LW. Cross-national price differences for pharmaceuticals: how large, and why? J
Health Econ 2000;19:159–95.
4 Public Hospital Accounts 1999. Paris: La Documentation Francaise, 2000.
5 The WHO Health Report 2000: Health Systems,
Improving Performance. Geneva: WHO, 2000.
6 Le Pen C, Rozenbaum W, Downs A, et al. Effect of
HAART on health status and hospital costs of
severe HIV-infected patients: a modeling approach.
HIV Clin Trials 2001;2:136–45.
7 Le Pen C, Rozenbaum W, Downs A, et al. A costeffectiveness analysis of changes in therapeutic strategies in the treatment of HIV since 1996. Therapie
2002;57:27–33.
8 Williams A. “Need” as a demand concept. In:
Culyer AJ, Wright KG, eds., Economic Aspects of
Health Services. Oxford: Martin Robertson, 1978.
9 Baudoin C, Rouland J, Le Pen C. Changes in Medical and Surgical Treatments of Glaucoma between
1997 and 2000 in France. 108th Congress of the
French Society of Ophthalmology, 2002.